Dynamic Assessment of Agriculture and Economic Growth Nexus in Morocco: Evidence from Structural VAR and Directed Acyclic Graphs

Abstract

The recurrence of international crises and their negative impact on the economy and household food security has stimulated a strong revival of interest in the role of the agricultural sector and its relationship with the national economy. Recently, a macro-econometric model has shown a well-established bidirectional causality nexus between the agricultural sector and the Moroccan economy. However, the assessment of the magnitude of effects in both directions and their historical evolution are crucial topics that have not yet been explored. The current study empirically examines the dynamic interrelationships between Moroccan agriculture and GDP using the structural VAR model. The data set consists of the annual macroeconomic time series covering the period 1980-2019, namely: GDP per capita, agricultural GDP, investment rate, money supply and trade openness. This paper exploits recent advances in artificial intelligence to determine the over-identifying restrictions, through Directed Acyclic Graphs. Impulse response functions reveal that the Moroccan economy is very sensitive to agricultural shocks compared to shocks due to other endogenous variables, meanwhile the agricultural sector is very reactive to its shocks. The results from the variance decomposition show that the agricultural shocks are the most important driver of economic growth fluctuations and account for almost 69% of the forecast error variance for the first year. The share of GDP shocks in the variance of the forecast error of agricultural GDP does not exceed 7% for a ten-year horizon, while agricultural shocks dominate the decomposition variance profile and never fall below the 74% threshold. These results highlight the predominance of the Agriculture-Led Growth hypothesis in comparison with the Growth-Led Agriculture hypothesis. The findings resulting from the historical decomposition reconfirm the historical dependence between the national economy and agriculture. This sector sometimes acts as a shock absorber, counteracting the poor performance of other sectors of the economy. Under the Structural VAR model, the historical analysis illustrates that the national economy is increasingly resilient to agricultural shocks because of the improved resilience of Moroccan agriculture to climate shocks. Although the impact of agriculture is historically prominent, the magnitude of its impact has significantly reduced by 22% between 1982-1999 and 2000-2019. Given the strong potential of the agricultural sector to promote economic growth, policymakers should continue to create favorable conditions to support the development of the sector.